A prospective study of outcome and predictors of subclinical and clinical depression in an older biracial sample of psychiatric outpatients

Objectives continues to be a debate about the long-term prognosis of psychiatric treatment of depression in later life. There have been no long-term naturalistic studies of psychiatric outpatient treatment of older adults in the United States. This study examines outcome and predictors of various levels of depression among a biracial sample of geropsychiatric outpatients in Brooklyn, NY. s ducted a naturalistic study (median: 33 months) of 143 persons aged 55 and over with diagnoses of depression drawn from 15 psychiatry outpatient clinics and 2 geriatric day programs. Their mean age was 68 years, 87% were female, and 43% were white and 57% were black, among whom 37% were African Caribbeans. Using Georgeʹs Social Antecedent Model of Depression, we examined the impact of 15 predictor variables on two outcome measures: presence of any either subclinical or clinical depression (CES-D score ≥ 8) and presence of clinical depression (CES-D score ≥ 16). s low-up, 84% and 90% of subclinically and clinically depressed persons at baseline, respectively, were depressed (CES-D ≥ 8); 62% of those in remission at baseline were depressed. In logistic regression, 3 variables were significant predictors of any level of depression on follow-up: baseline depression, baseline anxiety, greater increase in anxiety symptoms during the follow-up period. These 3 variables along with financial strain were significant predictors of clinical depression on follow-up. There were no inter- or intra-racial differences in outcome. sions eak outcome findings among older psychiatric outpatients in Brooklyn were consistent with unfavorable results reported in European studies. Because there were only a few predictors of outcome, strategies that target high risk persons – e.g., such as those with subclinical depression, anxiety, or in more economic distress – may prevent transition to severe and persistent depressive states.